← Back to community benchmarks
Qwen3.6-35B-A3B-oQ4-fp16-mtp
Performance
32k
tokens
765.9
PP tok/s
57.6
TG tok/s
42786
TTFT (ms)
23.0
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.4.4
macOS
macOS 26.5.2
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 721.2 | 70.8 | 19.9 GB | view |
| 4k | 965.1 | 68.5 | 20.7 GB | view |
| 8k | 969.6 | 66.0 | 21.1 GB | view |
| 16k | 904.3 | 58.2 | 21.7 GB | view |
| 32k | 765.9 | 57.6 | 23.0 GB | current |
| 64k | 513.3 | 45.2 | 25.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 70.8 | 1.00× |
| 2× | 94.4 | 1.33× |
| 4× | 120.0 | 1.69× |
| 8× | 160.8 | 2.27× |